Communications and Networking. 11th EAI international Conference, ChinaCom 2016 Chongqing, China, September 24-26, 2016, Proceedings, Part II

Research Article

Energy-Efficient and Latency-Aware Data Placement for Geo-Distributed Cloud Data Centers

Download
201 downloads
  • @INPROCEEDINGS{10.1007/978-3-319-66628-0_44,
        author={Yuqi Fan and Jie Chen and Lusheng Wang and Zongze Cao},
        title={Energy-Efficient and Latency-Aware Data Placement for Geo-Distributed Cloud Data Centers},
        proceedings={Communications and Networking. 11th EAI international Conference, ChinaCom 2016 Chongqing, China, September 24-26, 2016, Proceedings, Part II},
        proceedings_a={CHINACOM},
        year={2017},
        month={10},
        keywords={Energy-efficient Latency Energy consumption of servers Energy consumption of network transport Data placement},
        doi={10.1007/978-3-319-66628-0_44}
    }
    
  • Yuqi Fan
    Jie Chen
    Lusheng Wang
    Zongze Cao
    Year: 2017
    Energy-Efficient and Latency-Aware Data Placement for Geo-Distributed Cloud Data Centers
    CHINACOM
    Springer
    DOI: 10.1007/978-3-319-66628-0_44
Yuqi Fan1,*, Jie Chen1, Lusheng Wang1, Zongze Cao1
  • 1: Hefei University of Technology
*Contact email: yuqi.fan@hfut.edu.cn

Abstract

Cloud computing technology achieves enormous scale by routing service requests from users to geographically distributed servers, typically located at different data centers. On one hand, energy consumption of data centers and networks has been receiving increasing attention in recent years. On the other hand, users require low latency during data access from data centers. In this paper, we tackle the problem of energy-efficient data placement in data centers, taking into account access latency, energy consumption of data centers and network transport. We propose two request-routing algorithms to determine the number of copies for each data chunk and the data centers accommodating the data chunk. Our simulation results have shown that the proposed algorithms are effective in terms of the tradeoff among the data access latency, the energy consumed by network transport and data centers.